Supercomputers/Systems/Selene
USA · operational · ai training
Selene
Operated by NVIDIA at NVIDIA (undisclosed US site) .
Phase 1 seed dataset, compiled by hand. These rows were built from public operator, laboratory and vendor sources. A mechanical second-reader pass has since fetched every cited source: 32 of 52 systems have a readable citation that names them, and 4 are genuinely weakly sourced. Every claim carries its source and a confidence tier. Treat anything below verified as a lead, not a citation.
Measured performance
- Rmax
- 63.46 PFlop/s
- Rpeak
- 79.22 PFlop/s
- Rmax ÷ Rpeak
- 80.1%
- Cores
- 555,520
- Power
- 2.65 MW
- Per watt
- 24.0 GF/W
Figures are as last publicly reported for the configuration described below, not a live measurement. Where a system was upgraded in place, the post-upgrade configuration is shown and the earlier one appears in the timeline.
What this machine is made of
One row per supplier relationship. “Supplier at build” is the company that shipped the part at the time; where that company has since been acquired, the parent it rolls up to today is shown beside it. That distinction is what makes ticker-level aggregation possible across a thirty-year dataset.
| Role | Part | Supplier at build | Quantity | Confidence | Source |
|---|---|---|---|---|---|
| Integrator | - NVIDIA DGX A100 SuperPOD | NVIDIA | - | Reported | nvidia.com |
| Accelerator | NVIDIA A100 SXM4 40GB Accelerators NVIDIA A100 SXM4 40 GB | NVIDIA | - | Reported | nvidia.com |
| CPU | AMD EPYC 7742 (Rome) CPUs AMD EPYC 7742 64C 2.25GHz | AMD | - | Reported | nvidia.com |
| Interconnect | NVIDIA InfiniBand HDR Interconnect Mellanox HDR Infiniband | NVIDIA | - | Reported | nvidia.com |
| Cooling | Air cooling (CRAC/CRAH) Cooling DGX A100 SuperPOD: air cooled | NVIDIA | - | Reported | nvidia.com |
| Operating system | Ubuntu Server Operating systems Ubuntu | Canonical | - | Reported | nvidia.com |
The read
NVIDIA benchmarking its own reference architecture. Selene matters here as a calibration point rather than a science machine: it is one of the few DGX SuperPOD deployments measured under HPL, which gives the shadow-list estimator something real to anchor against when the commercial clusters built on the same design never report.
Timeline
Public list appearances
Pointers only: rank and edition, linking to the canonical entry. We do not reproduce list tables. See sourcing policy.
Site & facility
Source check
We fetched this system's own citations and recorded whether each page actually mentions it. This is a corroboration signal, not a fact check, and it is published so you can see how well the sourcing holds up rather than take it on trust.
None of the cited sources could be read from our network: the hosts returned bot challenges or did not respond. This check says nothing about the citations either way. See how this check works.
| Cited source | Result | Found on the page |
|---|---|---|
| nvidia.com | blocked our fetch | - |
Checked 2026-09-08 by pnpm verify. Re-run it and the table changes with the web.
Further reading
Inside a node
Why accelerators exist, what the memory hierarchy costs, and how the CPU and accelerator merged onto one package.
Power and cooling
Why air ran out at 20 kW a rack, how warm-water cooling removes the chillers, and what PUE conceals.
Operating a supercomputer
Acceptance testing, why "installed" and "in production" are six months apart, and the machine lifecycle.